Scribd
Upload
2K views

Traffic Sign Recognition - PPT

The document presents a preliminary project report for traffic sign recognition. The objectives are to identify and classify traffic signs from real-time video input and provide driver assistance by warning about upcoming signs. Different approaches like neural networks and color/shape-based template matching are discussed. The targets are to detect Indian traffic signs in real-time, possibly multiple signs, and implement the algorithm in OpenCV and on a TI processor for real-time demonstration by April 2012.

Uploaded by

Dheeraj Suvarna
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2K views

Traffic Sign Recognition - PPT

The document presents a preliminary project report for traffic sign recognition. The objectives are to identify and classify traffic signs from real-time video input and provide driver assistance by warning about upcoming signs. Different approaches like neural networks and color/shape-based template matching are discussed. The targets are to detect Indian traffic signs in real-time, possibly multiple signs, and implement the algorithm in OpenCV and on a TI processor for real-time demonstration by April 2012.

Uploaded by

Dheeraj Suvarna
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

TRAFFIC SIGN RECOGNITION

PRELIMINARY PROJECT REPORT Submitted by


Kishan Ganapathi Phalguni 08EC28

08EC46
Under the Guidance of Dr. Sumam David Mr Venugopal Madumbu Mrs Resmi Rajendran (TI)

(TI)

Objective
To identify and classify the traffic signals from a real

time video input. Driver assistance by warning the person about the traffic signs ahead

Motivation
Increase safety and prevent accidents. Significant step in introduction of automated driving

systems.

Background
Different methods neural networks, genetic

algorithms, simulated annealing, histographs, etc. Many simulations, few implementations. Research at Diamler Benz (Germany). Implementation using FPGA, multi-core SoC, etc. Accuracy = 65 to 96% Time for computation = 0.3s to 28s

Choice of Approach
Criteria: Simple, Fast, Accurate

Avoid neural networks as they require training.


Colour of signs - Distinguishing feature. Regular Shape of Signs - Circle, triangle and

rectangle. Euclidean distance measurements for template matching. Tracking (using Kalman filters) if necessary.

Targets
Detect and identify Indian traffic signs in real time.

Attempt to detect multiple signs.


Information boards with text are not considered. Identify the speed limits Implement the Algorithm in OpenCV. On Board Implementation, i.e. on TI processor

C64XX Real time demo.

Overall Functional Diagram

Targets
OpenCV code implementation - 2nd week of

October 2011 Implementation on C64X board - 2nd week of February 2012 Final Demo - 2nd week of April 2012

References
N. Kehtarnavaz and A. Ahmad, Traffic Sign

Recognition in Noisy Outdoor Scenes, IEEE Intelligent Vehicles Symposium (IV), Detroit, MI, USA, Sept. 1995, pp 460465. John Hatzidimos, Automatic Traffic Sign Recognition in Digital Images, Proceedings of the International Conference on Theory and Applications of Mathematics and Informatics ICTAMI 2004, Thessaloniki, Greece, pp. 174 184. Hasan Fleyeh and Taha Khan, Pattern Matching Approach Towards Real-time Traffic Sign Recognition, 2010 International Conference on Multimedia Computing and Information

You might also like